https://github.com/arkanto99/computer-vision
Projects of the subject Computer Vision at the University of Granada (UGR). Course 2021/2022
https://github.com/arkanto99/computer-vision
computer-vision convolutional-neural-networks keras
Last synced: 3 months ago
JSON representation
Projects of the subject Computer Vision at the University of Granada (UGR). Course 2021/2022
- Host: GitHub
- URL: https://github.com/arkanto99/computer-vision
- Owner: arkanto99
- Created: 2022-02-01T17:29:32.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2023-02-28T17:57:20.000Z (over 3 years ago)
- Last Synced: 2025-10-13T01:25:15.174Z (9 months ago)
- Topics: computer-vision, convolutional-neural-networks, keras
- Language: Jupyter Notebook
- Homepage:
- Size: 12.9 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Computer Vision Project
Projects of the subject Computer Vision at the University of Granada (UGR). Course 2021/2022
## Contents
### [Assignment 0](https://github.com/arkanto99/VisionPorComputador/tree/main/P0)
Image reading with OpenCV, Numpy and MatPlotLib.
### [Assignment 1](https://github.com/arkanto99/VisionPorComputador/tree/main/P1)
Manual implementation of different tools used in Computer Vision:
+ Approximation of Gaussian mask and its 1st and 2nd derivative masks.
+ Convolution using separable masks.
+ Construction of Gaussian and Laplacian pyramids.
In addition, two applications of the above are added:
+ Reconstruction of an image from its Laplacian pyramid.
+ Creation of hybrid images (greyscale and colour).
### [Assignment 2](https://github.com/arkanto99/VisionPorComputador/tree/main/P2)
+ Implementation of the SIFT algorithm for characteristic points detection
+ Construction of panoramas using Homographs
### [Assignment 3](https://github.com/arkanto99/VisionPorComputador/tree/main/P3)
Use of convolutional neural networks.
### [Final Proyect](https://github.com/arkanto99/VisionPorComputador/tree/main/Proyecto)
Use of the ResNet50 model to perform Transfer Learning
(using such a network previously trained on ImageNet) in order to assist in the diagnosis of pulmonary pneumonias in X-ray images. Other different models are also tested to discern which one provides the best results for this problem.
## Tech Stack
**Languages:** Python 3.7
**Libraries:** OpenCV, Numpy, Keras, Matplotlib
**Others**: Google Colab